wudongliang

you can use `--use-new-schema-change`. [flink-connector](https://doris.apache.org/zh-CN/docs/dev/ecosystem/flink-doris-connector?_highlight=flink#%E4%BD%BF%E7%94%A8flinkcdc%E6%8E%A5%E5%85%A5%E5%A4%9A%E8%A1%A8%E6%88%96%E6%95%B4%E5%BA%93%E6%94%AF%E6%8C%81mysqloraclepostgresqlsqlserver)

@icchux In this first schema change, in which scenario will the initialization field and the changed field be the same? Can you provide the corresponding table structure and detailed reproduction...

Hi, I followed the steps you mentioned exactly and did not reproduce the problem described above. @icchux

I guess the situation you described above is caused by the corresponding` extractDDLList` method being executed twice when you were debugging? But this will not happen in actual production.

Thank you for raising this issue. I confirmed that this problem does exist here. This problem occurs when the data in the table is empty.

Thanks for your contribution. In doris you can set the `lower_case_table_names` variable to ignore case, so providing this functionality in a sync tool is a bit redundant.

Is it possible to default to lowercase when writing, and then turn on this switch when uppercase is needed?

Since the community is global and there are many foreign developers, thank you for changing all descriptions to English.

run build all

run buildall